Tuesday, November 27, 2007

What model?

Ah, don't we love medicine and all its supporting science and engineering??

Earlier posts harped about our love of abstractions and about our potential misuses of such.

The last post brought up the idea that morbidity & mortality could be applied to business analysis; actually, implied is that we need some metaphor like this, with a natural basis, to resolve issues of quasi-empiricism.

Well, the idea of (abstraction suggested by) this post may have some power.

Such as, just as a product deficiency may point to some morbid factor in the related process leading back, of course, to the organization(s) and the players thereof, so too can some economic deficiency in that major of organizations, the society, point to some morbidity issue that we ought to understand.

This model might help differentiate how truth engineering, 7 'oops 7, and whosenoseknows might interplay, namely, the theoretical/abstract basis, the operational issues, and measurement, respectively.

Monday, November 26, 2007

Morbidity, mortality

Every week the medical establishment reports on their incidents of morbidity & mortality (ah, we ought to love the M&M) to the CDC. The former relates to what has been seen in terms of illness; the latter relates deaths and probable causes.

But, too, every good medical organization does reviews for their own benefit. Here is an interesting collection that we'll go through looking for parallels with business reviews.

That we see errors in the medical community is not a problem; oops happen when we have 'oops and loops. It's the reporting and analysis that we ought to find commendable (as many of us do); business seems to want to hide oops, though they ought to know that oops are more probable than not when involved with 'oops and loops.

In business, does only the good news flows up? Actually, the process, from the top to the down, ought to ask every subordinate at a particular level to talk one problem in more detail than may be usual; that would be a regular 'oop; then the loop would be to push this down the spiral (helix as organization structure); could we identify oops earlier in their evolution?

Now morbidity has to do with the health of the organization and its resources (and processes). It may be that health is too much assumed; perhaps, too, that is where the quasi-empirical tie comes in; that is, health would be indicated somehow, quantitatively and not, in reports, such as a dashboard; how well-founded, or strong, are those indicators would be one question to always consider; after all, business can (perhaps, ought to) learn to use scientific methods.

Yet, assuming health, then there is the function of the organization; if it is product design and development, then the product would be evaluated; it may be that product deficiencies point to process illness (bringing up, then , morbidity); even if it turns out that some material issue is the culprit, still the healthy process ought to cause identification of the problem, perhaps prior to things getting too awry.

Modified: 11/27/2007

Saturday, November 24, 2007

'Oops and loops lead to oops

Even doing nothing can have consequences which might be interesting to look at.

However, we want to look at efforts of doers that have results which imply then that we'll have some 'oops through which the doers need to leap as well as the loops involved with doing it again (any good learning circuit involves feedback and knowing when to quit).

Of course, that oops can follow is to be expected. How well these are managed is a key factor.

Gosh, just parsing through this post so far offers up many nuances that could be explored in terms of the former focus, but we ought to let that lie for the time being. There are many other useful twists to turn.

An example might be driving given some habits that have come to fore with the gaming generation. How many oops occur from inattentiveness of various sorts?

Here are three, of many, recent possible examples of problematic driving.
  • going along a familiar route, at an intersection, turn by rote, that is, attention of the uper cortex elsewhere, driving being done in a reactive mode, not observing, finding that there is a blockage in the lane, thankfully being aware enough to stop, how many times is the reaction time too long for safety?
  • driving with divided attention, due to too little thought being given to processing the data related to the driving act (text-messaging, etc.), again, in some situations, this can have serious consequences
  • not accounting for possible moves by others, such as speeding pass some cars in a queue in a lane, completely ignoring what has motivated the defensive driving moral
Though a route may have been followed many times, each time through is not a replay of the former times. The data related to the current route needs to be processed and handled. Applying rote responses can be problematic, even though, for a whole class of needed actions, we use what has been learned.

As, we can always ask: how often are the situations so novel as to require creative means? Well, one retort would be that it would be first time through.

This little thought could be applied to planning, especially in terms of earned value. Analytics ought not be weighted too heavily the first time through, success with parametrics notwithstanding. It may be that the thread that runs through time (collapse of the graph on the critical path, so to speak) is such that only routine methods are necessary.

Yet, on the first pass, even the first several, one might think that markers and milestones themselves ought to be scrutinized. Nothing would be entirely routine in this scheme.

Of course, driving like that would be extremely tedious. We want speed (and spin).

After thought: can the computer's role as simulator (of which there are several types and uses) remove (or is it reduce?) the necessity of proof in reality?

Remarks:

03/25/2013 -- The Atlantic had an article about King Abdullah II. Now, he is an example of a doer, from several angles. What I liked when I read it was that while being educated in Massachusetts, he bussed tables. What that means for those who don't know is clean up dirty dishes and such. When I, as a young man, was in the US Army, we had still had KP duty which included such types of things. Another task that ought to be tried once by everyone: cleaning the grease pit.

12/14/2010 -- This post deals with, while not explicitly stating so, issues related to compiled knowledge.

Modified: 03/25/2013

Wednesday, November 21, 2007

Thanksgiving

One thing that we can be thankful for is that, for the most part, thoughts (or expectations) of easy accomplishment in difficult situations, or of nice solutions to difficult problems, are not rewarded, at least not to the degree we would like.

For those wishes that are rewarded, even if only to some degree and as probably happens more than might be good for us, someone covered for us, about which we ought to be grateful; until our maturity, many played a support role; in maturity, expertise, especially as provided by humans (as resources), abounds in every direction (much of this may be taken without thanks - or is misused - it sure looks that way to the insightful observer); so many examples could apply.

The gilded cage, related to the 'golden touch' (in the modern era, mostly attained through exploiting the substance of many others) is not as wonderful inside as it might look to the onlooker. The better cage gets involved with learning from mistakes so as to push back our known frontiers into useful territory.

Just as flightblogger changed course, so, too, will this blog. The name will be the same as the motivating influence can serve as one, of many, sources for examples to use to describe the issues that are involved in several modern dilemmas, as well as to define alternative attempts at coping with such issues with fair, equitable, truthful, and effective means.

Remarks:

01/18/2009 - We even need to look at why we need finance.

12/03/2008 -- We need to learn what we might be taught about money by Islamic Finance.

11/26/2008 -- The mess grew and grew, fairy dusting indeed.

07/31/2008 -- It's not enough to rant and spout off. So, let's start something constructive by looking at money and what it is.

Modified: 01/18/2009

Tuesday, November 20, 2007

Poll completion 3


Of the 20 voters in the third poll, most of whom were Suppliers (6, 54%) according to the Role poll, 13 (65%) indicated 2009 as the most likely delivery year. The next most likely year was 2010 (5, 25%).

A tongue-in-cheek 'Before the A350' got 1 vote.

Of those who did the Role poll, the next highest group were the flying afficionados (4, 33%).

Prior polls: First, second.

Modified: 01/11/2008

Friday, November 9, 2007

Third poll series

Note: Poll completion.
Announcements:


- You can support your favorite charity by using bet2give (787 in 2008).
- Also, flightblogger is back; we'll see what information comes from that source; the context will change due to the flightglobal connection.

The poll:

The first poll and second poll are completed. This third poll is a short one to start to get some feedback on opinions about when the 787 might be delivered.

- Based upon the current information that we can know, when will the 787 be delivered? -- We, outside of OEM, only have limited information with which to make a determination about this, however there are many rumors floating around that cover a spectrum: will things go forward nicely such that the product will deliver with only a minor delay (2009/10)? Or, are there more problematic issues that will be difficult to overcome (much later)?

- My role is with -- We won't be able to establish any correlations, however knowing who might be taking the polls might be of interest. We'll continue to have this part of a series.

Disclaimer:
Usual poll etiquette assumed; polls are oriented toward information and not mis / dis-information.
- A casual user cannot double vote on any poll. But, there is no guard against intentional duplicate votes by those who know how.
- There is no consistency checking between polls.
- There is no meta-information about who votes or why.
- There is no way to correlate between the polls, however the 'role' poll allows some indication of interest.

Modified: 01/20/2013

Friday, November 2, 2007

Carts need horses

Sometimes a production process proves to be problematic even in the best of situations with products that have been made before.

With a new product, can one expect to spread production before the concept has been proven? Even with the best of analytics, could this be somewhat like the old adage of putting the 'cart before the horse'?

There are development issues to consider, too, especially with all the new capabilities that enable and lead toward efforts at a geographically-dispersed team.

The distributed model might expect suppliers to become expert quickly in a new area; attainment of expertise requires time (and one might add repetition, continued improvement, and other practices associated with the best efforts).

Farmed-out work has a worker and looker flavor needing insightful overview, hopefully with more than just the critic's role for the looker. Expertise is not a vicariously obtained attribute.

Too, would prudence suggest that production might need to lag behind progress on the proof axis?

In the past, the effort to create a physical product model helped the 'proof' advancement; with the onset of computer modeling, the question arises of how much the model may replace reality. Much about that question is still an open issue; when the product is complex and has stringent performance criteria, ought we expect a stronger temporal tie between analytical tests and physical tests throughout the process?

Remarks:

01/19/2011 -- Update1 and Update2. The focus now will be mostly the idiots of economics/finance.

09/02/2009 -- Lets face it, folks, undecidability needs to be discussed and adopted in any complex situational setting, especially if computers are involved. Only hubris pushes us to make loud exclamations about what we're going to do in the future.

07/14/2009 -- Nope, confounding continues.

05/18/2009 -- Testing in flight is within sight.

01/28/2009 -- A few related posts that came later: IQ and PIQ, Earned value, Hype cycle, Outsourcing as panacea. Expect a recap on this theme.

11/26/2008 -- Problems continued to arise in 2008. As of now, some static tests have been done. But, issues with supply management were troublesome. Of course, an IAM strike caused a little delay. But, there has been no test flight, as of yet, so functional issues remain more unknown than not. See Polls for an idea of opinion at various points.

Modified: 01/19/2011